data/llm_text_generation/code/H100 80GB HBM3/codellama/CodeLlama-13b-hf/bs192+tp1+pp1.json

{
  "Model": "codellama/CodeLlama-13b-hf",
  "GPU": "NVIDIA H100 80GB HBM3",
  "TP": 1,
  "PP": 1,
  "Energy/req (J)": 26.030655990596767,
  "Avg TPOT (s)": 0.09005491217306259,
  "Token tput (tok/s)": 1453.2242990888508,
  "Avg Output Tokens": 96.02926829268293,
  "Avg BS (reqs)": 183.62448979591838,
  "Max BS (reqs)": 192
}